Automated control of individual droplets in microfluidic channels offers tremendous potential for applications requiring high accuracy and minimal user involvement. The feasibility of active droplet control has been previously demonstrated with pressure-driven flow control and visual feedback, but the manual operation required to perform droplet manipulations limited the accuracy, repeatability, and throughput. The present study improves upon the aforementioned challenges with a higher-level algorithm capturing the dynamics of droplet motion for a semi-automated control system. With a simple T junction geometry, droplets can now be automatically and precisely controlled on-demand. Specifically, there is ±10% accuracy for droplet generation, ±1.3% monodispersity for 500 μm long droplets and ±4% accuracy for splitting ratios. On-demand merging, mixing, and sorting are also demonstrated as well as the application of a drug screening assay related to neurodegenerative disorders. Overall, this system serves as a foundation for a fully automated system that does not require valves, embedded electrodes, or complex multi-layer fabrication.

Accurate estimation of lithium-ion battery (LIB) state of health (SOH) underpins safe operation, predictive maintenance, and lifetime-aware energy management. Despite recent advances in machine learning (ML), systematic benchmarking across heterogeneous real-world cells remains limited, often confounded by data leakage and inconsistent validation. Here, we establish a leakage-averse, cross-battery evaluation framework encompassing 32 commercial LIBs (B5–B56) spanning diverse cycling histories and temperatures (≈4 °C, 24 °C, 43 °C). Models ranging from classical regressors to ensemble trees and deep sequence architectures were assessed under blocked 5-fold GroupKFold splits using RMSE, MAE, R² with confidence intervals, and inference latency. The results reveal distinct stratification among model families. Sequence-based architectures—CNN–LSTM, GRU, and LSTM—consistently achieved the highest accuracy (mean RMSE ≈ 0.006; per-cell R² up to 0.996), demonstrating strong generalization across regimes. Gradient-boosted ensembles such as LightGBM and CatBoost delivered competitive mid-tier accuracy (RMSE ≈ 0.012–0.015) yet unrivaled computational efficiency (≈0.001–0.003 ms), confirming their suitability for embedded applications. Transformer-based hybrids underperformed, while approximately one-third of cells exhibited elevated errors linked to noise or regime shifts, underscoring the necessity of rigorous evaluation design. Collectively, these findings establish clear deployment guidelines: CNN–LSTM and GRU are recommended where robustness and accuracy are paramount (cloud and edge analytics), while LightGBM and CatBoost offer optimal latency–efficiency trade-offs for embedded controllers. Beyond model choice, the study highlights data curation and leakage-averse validation as critical enablers for transferable and reliable SOH estimation. This benchmarking framework provides a robust foundation for future integration of ML models into real-world battery management systems.